Pressure sensors made by micro machining methods are well known and considered one of the most mature applications for MEMS technology. Since the early 1970's, pressure sensitive diaphragms have been formed from silicon substrates, the deflection of which have been detected by optical, piezoresistive, piezoelectric or capacitive means. So far, the most significant detection method used for commercial applications has been piezoresistive detection, which is convenient to implement since single crystal silicon is an inherently piezoresistive material. Examples of piezoresistive pressure sensors are disclosed in U.S. Pat. Nos. 3,893,228, 3,916,365, 4,203,327, and 4,763,098.
Another significant method is capacitive detection, which provides for lower transducer noise and better thermal stability, but requires more complex mechanical structures, since the capacitance between the movable diaphragm and a fixed counter electrode must be established. Examples of capacitive pressure transducers are disclosed in U.S. Pat. Nos. 4,257,274, 4,881,410, 4,625,561 and 5,936,164. An important realization for remote sensing purposes is that capacitive transducer devices do not consume power, as is the case for piezoresistive devices in which a biasing resistor must be used to detect a change in voltage or current. In remote sensing it is desirable to minimize transducer power consumption to reduce the size of the required power source (i.e., battery). If a capacitive transducer is combined with a coil, an LC circuit with theoretical resonance frequency of fres=(2π√{square root over (LC)})−1 is formed. If the coil is further designed, such that an external electromagnetic field may easily be coupled into the coil (i.e., a planar coil), the resonance frequency of the LC circuit may be detected remotely by analyzing the coupling impedance of the LC circuit to a transmitter coil. A pressure induced change of capacitance C in the transducer then leads to a shift in the LC circuit's resonance frequency, which may be detected remotely. Wireless pressure transducers based on this approach are disclosed in L. Rosengren et al., “A system for passive implantable pressure sensors”, Sensors & Actuators, vol. A43 (1994), pp. 55–58 and in U.S. Pat. No. 6,287,256.
A prior art wireless pressure sensor 10 is shown in FIG. 1. A silicon substrate 2 is etched from both sides to form a recessed diaphragm 3 and cavities 6. On a separate glass substrate 1, a planar metal inductor coil 9 is formed with windings 7. Also formed on glass substrate 1 are a fixed counter electrode 5 and an electrical connection 8. The silicon substrate 2 and glass substrate 1 are bonded together using anodic bonding methods to form the complete pressure sensor 10. When bonded together, the recess at the diaphragm 3 establishes an operational air gap 4 between the diaphragm 3 and the fixed counter electrode 5. An important parameter for the inductor coil 9 used in conjunction with the capacitor is the quality factor (Q), which is a measure of the sharpness of the resonance, and hence the relation between inductance and resistive loss of the electrical connection 8 and coil 9. The quality factor directly influences the precision with which the resonance frequency can be determined by inductive coupling, and therefore, the resolution of the pressure sensor 10. Unfortunately, in prior art devices based on planar coils, as shown in FIG. 1, there are several limitations that affect the quality factor of the coil. First, the number of windings 7 that can be realized is restricted, since they are placed outside the diaphragm 3 and therefore, add to the overall dimensions of sensor 10. Second, the materials used to form the windings 7 of the coil 9 are typically deposited by electroplating to achieve sufficient metal thickness. Electroplated metals are known to have inferior resistivity compared to metals deposited by other means, which therefore results in significant resistive losses in the coil 9.